Profile My skills Saved Collections Edit profile Submit a skill

← all publishers

gn00678465

@gn00678465 source repo

10 published skills

  1. ▌
    Teach · gn00678465 bundle
    Teach the user a new skill or concept, within this workspace.
    0
    installs
  2. ▌
    Crawler · gn00678465 bundle
    Fetches web pages and converts them to clean markdown using a robust 3-tier chain (Firecrawl → Jina Reader → Scrapling stealth browser). Use this skill instead of WebFetch whenever the user provides a URL and needs the page's text content — especially for sites that block direct access: medium.com articles (paywalled/metered), WeChat public accounts (mp.weixin.qq.com, geo-restricted), documentation sites with bot protection, or any page where simple HTTP fetching might return a CAPTCHA or empty page. Triggers for: "read this URL", "summarize this article/page", "grab the content from", "extract text from", "what does this page say", "fetch this link", or any request to access and process a specific web page. Do NOT trigger for: building scrapers, checking HTTP status codes, parsing already-downloaded HTML files, answering conceptual questions about scraping tools, or monitoring page changes.
    0
    installs
  3. ▌
    Meta Ralph · gn00678465 bundle
    Scaffold (bootstrap) OR amend a ralph autonomous coding loop in a target git repo — writes prd.json + .ralph/prompt.md + .ralph/ralph.<sh|ts|js|py> + RUNBOOK + .gitignore. Pure scaffolder; does NOT execute the loop. Use for first-time setup OR to append more user stories to an existing scaffolded prd.json. Bootstrap trigger phrases: "init ralph", "set up ralph", "scaffold ralph", "bootstrap ralph", "建立 ralph", "初始化 ralph", "ralph 起手". Amend trigger phrases (every phrase MUST carry ralph/prd/meta-ralph context): "append stories to ralph", "add user stories to ralph prd", "extend the ralph prd", "extend ralph backlog", "grow the ralph prd", "在現有 ralph prd 補 stories", "追加 ralph stories", "在 ralph 加 user story", "ralph prd 新增 story". NOT for running an existing ralph loop (use ralph-loop:ralph-loop instead), NOT for explaining what ralph is (informational queries don't need this skill), and NOT for generic PRD editing, backlog grooming, or writing user stories outside an existing meta-ralph scaffold.
    0
    installs
  4. ▌
    Skill Forge · gn00678465 bundle
    End-to-end workflow for forging and updating agent skills that match user intent, stay functional, and minimize token usage. Routes between create and update commands; orchestrates grill-me (clarify), write-a-skill (draft), skill-creator (eval/validate), and skill-cleaner (optimize). Use when the user wants to create, update, or validate a skill end-to-end, or mentions "skill forge".
    0
    installs
  5. ▌
    Skill Cleaner · gn00678465 bundle
    Audit Codex/OpenClaw skills: loaded roots, duplicate skills, unused skills, prompt-budget costs, compact descriptions.
    0
    installs
  6. ▌
    React AI Infra · gn00678465 bundle
    Deploys cross-platform agent assets into an already-created React project. ACTIVATE when a React/Next.js/TanStack Start/Vite repo (existing or freshly bootstrapped by its own CLI) needs APM-managed dependencies, bundled skills/rules, and a framework-specific AGENTS.md wired in for Claude Code, GitHub Copilot, and the cross-client `.agents/skills/` bus. Asks which React framework is in use, generates apm.yml at the project root, runs `apm install`, and writes assets to .claude/skills/, .agents/skills/, .claude/rules/, .github/instructions/, and the project root. Does NOT create the React app itself.
    0
    installs
  7. ▌
    Security Supply Chain · gn00678465 bundle
    Hardens npm/pnpm/bun/yarn/uv/pip configs against supply chain attacks via minimum release age gates, lifecycle-script allowlists, OIDC trusted publishing, and commit-time secret scanning. Use when auditing package manager configs, configuring dependency bots (Renovate/Dependabot), evaluating whether a fresh package version is safe to install, or responding to a published npm/PyPI compromise (chalk/debug, ua-parser-js, TanStack-style hijacks).
    0
    installs
  8. ▌
    Svelte AI Infra · gn00678465 bundle
    Scaffold a Svelte/SvelteKit repo's AI infrastructure source files — APM config (`apm.yml`), `.apm/` agents & instructions, `.lsp.json` — pinned to a fixed `sveltejs/ai-tools` SHA, ready for the user to run `apm install`. Use when asked to enable / initialize / set up Svelte AI tooling for a project, bootstrap svelte ai infra, or wire up sveltejs/ai-tools plugin. Do NOT use when only modifying a single skill file, a single target's config, .mcp.json alone, or apm.yml content (use Read+Edit directly).
    0
    installs
  9. ▌
    Improve Codebase Architecture · gn00678465 bundle
    Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.
    0
    installs
  10. ▌
    No Use Effect · gn00678465
    Enforce the no-useEffect rule when writing or reviewing React code. ACTIVATE when writing React components, refactoring existing useEffect calls, reviewing PRs with useEffect, or when an agent adds useEffect "just in case." Provides the five replacement patterns and the useMountEffect escape hatch.
    0
    installs